WO2006041140A1

WO2006041140A1 - 3d measurement sensor

Info

Publication number: WO2006041140A1
Application number: PCT/JP2005/018889
Authority: WO
Inventors: Yukio Sato; Susumu Shibata
Original assignee: Japan Science And Technology Agency; Spacevision, Inc.
Priority date: 2004-10-15
Filing date: 2005-10-13
Publication date: 2006-04-20
Also published as: JP2006112978A; US7560680B2; JP3774225B2; US20090045326A1

Abstract

In a matrix of photo-detectors (10) constituting a 3D measurement sensor, outputs of photo-sensors belonging to the same row are connected to the same signal line (12). Outputs of the photo-detectors (10) are read successively from left to right to a signal line (12) while being scanned by a horizontal scan circuit (14) during a short horizontal scan period of such an order that the slit light of the light cutting method can be considered to be substantially in a stop state. A position detection unit (18) determines the time from the moment of start of the horizontal scan period to detection of the maximum peak of the signal on the signal line (12). This time indicates the distance from the end of the scan start side of the row to the photo-detector (10) which has detected the peak of the slit light. A recording unit (22) records the value of the distance obtained by the position detection unit (18) while correlating it with the value of the counter (20) indicating the projection angle of the slit light.

Description

Specification

3D measurement sensor

Technical field

The present invention relates to a sensor for measuring a three-dimensional shape of an object.

Background art

[0002] Shape measurement of a three-dimensional object using an optical system is being used in various industrial fields. Among them, there is a light cutting method as a method considered to be highly practical. In the light cutting method, the reflected light on the surface of the object when the object is irradiated with slit-shaped or spot-shaped light is captured by the camera, and each point on the surface of the object is determined by the principle of triangulation based on the positional relationship between the light source and the camera. Find the three-dimensional coordinates of.

[0003] As an apparatus for measuring a three-dimensional shape by the optical cutting method, for example, there is an apparatus shown in Non-Patent Document 1 (see particularly "3.2.3 Image Encoder"). In this device, an image obtained when the slit light is swept over an object is captured by a CCD camera, and a video signal output from the camera is input to an image encoder. In addition to the video signal, the image encoder receives a code signal indicating the projection angle of the slit light. The image encoder performs luminance peak hold processing for each pixel in real time on the input video signal, and detects the timing at which each pixel takes the maximum luminance by the peak hold processing. By storing the code signal at, as a coded value of the pixel, a code image that can be applied to the light cutting method is formed.

However, the apparatus shown in Non-Patent Document 1 is based on the premise that the projection position of the slit light does not substantially change during one readout scanning period of the CCD camera. The readout scanning must be repeated while gradually changing the position of the slit light. Since the readout scan of a general CCD camera is about 1Z30 seconds or 1Z60 seconds, it takes a very long time to obtain the three-dimensional shape of one object with this device, and it is practically not applicable to moving objects.

[0005] To solve such a problem, one of the inventors of the present invention has proposed an apparatus shown in Patent Document 1. This device is equipped with a non-scanning image sensor with an element memory for each photodetector. Each element memory is connected to a bus carrying a time lapse signal t. Then, the light receiving output generated when the light detector detects the slit light is given as a trigger to the element memory, whereby the element memory is also supplied with the bus force and latches the value of the time lapse signal t. According to this configuration, while the slit light is swept once with respect to the object, information corresponding to the time when the slit light hits the corresponding photodetector is stored in each element memory of the image sensor. The That is, in this apparatus, information equivalent to the code image generated by the image encoder of Non-Patent Document 1 is accumulated in the element memory group of the image sensor during one slit light sweep. Therefore, it is only necessary to perform reading scanning of the memory group of the image sensor once per slit light sweep. As described above, according to the apparatus of Patent Document 1, shape measurement can be performed even when the slit light is swept at a speed equivalent to the reading scanning of the image sensor, so that shape measurement near real time can be performed.

[0006] However, in the apparatus of Patent Document 1, it is necessary to provide an element memory for each photodetector of the image sensor, which leads to an increase in the size of the image sensor or a disadvantage in terms of manufacturing. There was a problem such as becoming.

[0007] Further, Non-Patent Document 2 shows a conventional technique that takes another approach to solve the above problem. This is intended to enhance the real-time capability by reading out the image sensor at high speed. In addition to the photodetector array, a signal for four frames for each pixel (one pixel with four color photodetectors). A CMOS sensor having an analog memory array capable of storing the memory and a comparator and an output latch for each column of the memory array is used. In this sensor, the signal of each pixel is read out at a high frame rate of 3.3 kfps (kilo 'frame per second), and the read signal is stored in the cell of the corresponding frame of the analog memory array. Then, in order for each pixel in each column of the array, the sum of the previous two frames from the sum of the later two frames of the four frames of signals stored in the analog memory array by the comparator of that column. When this signal is 0, the output latch holds the value `` 0 '', and when the signal has a negative value of 0 (this is the peak of the slit light on the pixel) The timing is held at “1” in the output latch. If the output latch of each column is read out at high speed, it is possible to determine whether or not the slit light peak is applied to each pixel every frame (one readout scanning period of the photodetector array). The frame number is slit This sensor can determine the frame number at which each pixel detects the slit light peak, so that the three-dimensional shape of the object can be obtained.

[0008] The sensor of Non-Patent Document 2 is expected to improve the aperture ratio of the photodetector array by providing circuit configurations such as an analog memory array, a comparator, and an output latch outside the photodetector array. However, the analog signal memory for 4 frames per pixel is necessary, and there is a problem that the circuit scale of the entire device becomes large. Even if 4 frames per pixel is an example, if the peak of the slit light is obtained from the temporal difference of the received light signal of the same pixel, the principle is adopted. Analog signal memory for the frame is necessary.

In Patent Document 2, as an imaging device for three-dimensional measurement, a plurality of pixels are two-dimensionally arranged, and each pixel force has a shape that is long in the sweep direction on the imaging surface of S-slit light. In addition, a pixel in which pixels adjacent in a direction orthogonal to the sweep direction are shifted in the sweep direction is disclosed. In addition, this document compares the output values of adjacent pixels and memorizes the spatial position on the image sensor where the difference is inverted, which slit light hits which pixel! A method to identify whether or not is disclosed! (See paragraphs 7 and 26).

[0010] However, Patent Document 2 does not show a specific hardware circuit configuration for realizing the method.

[0011] Further, in Patent Document 3, as a device for obtaining a three-dimensional shape of an object, a zero-order moment calculator and a first-moment calculator calculate at each position in the direction in which the light pattern light extends. An apparatus is disclosed in which a zeroth-order moment and a first-order moment in a direction perpendicular to the direction are obtained by integrating digital signals corresponding to output signals obtained at substantially the same time by a light receiving element array. In this device, the photoelectric output of each light receiving element of the photodetector array is converted into a multivalued digital signal, and moment calculation processing is performed on the digital signal.

[0012] The device of Patent Document 3 requires a complicated circuit configuration for various processes such as analog-to-digital conversion of the photoelectric output of each light receiving element and parallel reading of the obtained digital value. The manufacturing cost was high.

[0013] Non-Patent Document 1: Toru Yoshizawa, “Three Dimensional Optics, One Optical Three Dimensional Measurement”, First Edition, New Technology Co., Ltd. The Fusions, March 1993, p. 38- 51

Non-Patent Document 2: Toshinobu Sugiyama, 3 others, “C MOS sensor capable of color video imaging and real-time 3D measurement”, The Institute of Image Information and Television Engineers Technical Report, The Institute of Image Information Media, 2002 3 May 18, 26th, 26th, p. 1-6

Patent Document 1: Japanese Patent Publication No. 6-025653

Patent Document 2: JP 2001-053261 A

Patent Document 3: Japanese Patent Laid-Open No. 2002-365022

Disclosure of the invention

[0014] The present invention provides a sensor capable of collecting data used for measuring a three-dimensional shape of a target with a relatively simple circuit configuration.

[0015] A three-dimensional measurement sensor according to the present invention is a sensor that acquires information for calculating a three-dimensional shape by a light cutting method, and is a photodetector configured by arranging photodetectors in a matrix. Select the array and the row signal line provided for each row of the photodetector array and connected to the output of each photodetector belonging to the corresponding row, and the photodetector of the photodetector array by column. The output signal of each of the photodetectors is output to the corresponding row signal line, and the image on the photodetector array of the slit light scanned for the light cutting method is substantially stationary. A row scanning circuit that switches a selected column in order from one end of the row to the other during a row scanning period that is short enough to be considered, and a light receiving position detection circuit that is provided for each row signal line. Each time, the corresponding row signal line flows A light receiving position detection circuit that detects a peak of the signal in the row scanning period and obtains information on a light receiving position in the row of the photodetector array in the row scanning period based on a timing at which the peak is detected; And an output circuit that outputs information on the light receiving position obtained by each light receiving position detection circuit for each scanning period.

Here, the selection of “in units of columns” of the photodetectors in the row scanning circuit is not limited to one column, but a plurality of consecutive columns may be selected together.

In one aspect of the present invention, each light receiving position detection circuit holds a peak hold circuit that holds a peak level of a signal on the corresponding row signal line, and a differential that differentiates an output of the peak hold circuit. A circuit, an integration circuit for integrating a predetermined reference signal, and the differentiation circuit A control element that discharges the charge accumulated in the integration circuit when the output of the circuit is turned on, and accumulates the charge in the integration circuit while the output of the differentiation circuit is at the off level. The output circuit outputs the output of the integrating circuit at the completion of the row scanning period as information on the light receiving position.

In another aspect of the present invention, the light receiving position detection circuit holds a peak hold circuit that holds the peak level of the signal on the corresponding row signal line, and a differentiation circuit that differentiates the output of the peak hold circuit. And a light receiving position signal generating circuit for generating a light receiving position signal having a signal level corresponding to the elapsed time of the starting force of the row scanning period, and the light receiving position signal at the time of the trigger, triggered by the output of the differentiating circuit. A sample hold circuit that samples and holds the signal, and the output circuit outputs a signal held by the sample hold circuit when the row scanning period is completed as information on the light receiving position.

In yet another aspect, the three-dimensional measurement sensor includes an angle counter for obtaining a projection angle of the slit light with respect to the measurement target, information on the light reception position obtained by each light reception position detection circuit during the row scanning period, and A storage processing unit for storing the output value of the peak hold circuit when the row scanning period is completed in association with the count value of the angle counter in the row scanning period;

[0020] The three-dimensional shape sensor of this aspect can provide information on the value of the angle counter and the light receiving position as data that is the basis for the calculation of the three-dimensional coordinates based on the principle of trigonometry. Furthermore, by combining the light receiving position information provided by this sensor with the output value of the peak hold circuit, a luminance image showing the luminance distribution of the measurement object can be formed. Brief Description of Drawings

FIG. 1 is a diagram for explaining the principle of three-dimensional shape measurement by a light cutting method.

FIG. 2 is a diagram for explaining the measurement principle of the three-dimensional measurement sensor in the embodiment.

FIG. 3 is a diagram showing a circuit configuration example of a three-dimensional measurement sensor in the embodiment.

FIG. 4 is a diagram showing a circuit configuration example of a peak hold circuit and a timing detection circuit.

5 is a diagram showing a timing chart of signals in each part of the circuit configuration example of FIG. 3.

FIG. 6 is a diagram showing a system configuration example of a three-dimensional shape measuring apparatus using the sensor of the embodiment. FIG. 7 is a diagram showing a main part of another example of the circuit configuration of the three-dimensional measurement sensor.

FIG. 8 is a diagram showing a main part of still another example of the circuit configuration of the three-dimensional measurement sensor.

FIG. 9 is a diagram showing a main part of still another example of the circuit configuration of the three-dimensional measurement sensor.

FIG. 10 is a diagram showing a main part of still another example of the circuit configuration of the three-dimensional measurement sensor.

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0023] First, the principle of three-dimensional shape measurement by the light cutting method to which the three-dimensional measurement sensor according to the present invention is applied will be described with reference to FIG.

[0024] As shown in Fig. 1, in the light cutting method, slit light 5 is emitted from a laser 1 and a lens system 2 with equal power, and the slit light 5 is reflected by a deflecting device 3 such as a polygon mirror. Target Project onto object 4. Thereby, the surface of the target object 4 is illuminated in a slit shape. The reflected light from the surface of the target object 4 is focused by the lens system of the imaging device 7 and forms an image on the imaging surface 8 of the imaging device. On the imaging surface 8, an image is formed in which only the slit-like portion 9 corresponding to the slit-like illuminated portion 6 on the surface of the target object 4 is bright and the other portions are dark. At this time, the three-dimensional coordinates of the point on the illuminated part 6 on the object 4 are the two-dimensional position of the point on the slit-like part 9 on the imaging surface 8 corresponding to the point, the light source (laser 1 ) And the position of the imaging surface 8 and the projection direction (angle) of the slit light 5 at that time, it is obtained by the principle of triangulation. Then, by rotating the reflecting surface of the deflecting device 3, the surface of the target object 4 is scanned with the slit light 5 to one end force to the other end. In principle, all the points on the surface of the target object 4 (except of course the part that becomes a shadow on the structure of the optical system) are illuminated by the slit light only once during this one scan, so the triangulation mentioned above The three-dimensional position can be specified by the principle of.

[0025] Although the outline has been described above, the principle of the optical cutting method and the rough system configuration are described in the above-mentioned prior art documents (especially non-patent document 1), Seiji Iguchi and Kosuke Sato, "Three-dimensional image measurement" “First edition, Shosodo, published on November 20, 1990, p36-40 (2.2.2 Slit light projection method), and is well known, so further explanation is omitted.

In the present embodiment, a three-dimensional measurement sensor suitable for three-dimensional shape measurement by this optical cutting method is provided. The measurement principle of this embodiment will be described with reference to FIG.

In the apparatus of this embodiment, slit light is incident on the position of each pixel (photodetector) on the imaging surface 8. Data of the specified time is generated. The time when the slit light is incident on the pixel indicates the projection angle of the slit light. Since the position of each pixel on the imaging surface 8 and the position of the light source are known, if the time is known, the three-dimensional coordinates of the point on the target object 4 corresponding to that pixel can be calculated. This concept itself is the same as in Patent Document 1. In the method of Patent Document 1, there is a problem in terms of cost and other points that a circuit for detecting the incidence of slit light and holding the time is provided for each pixel, and this embodiment aims to improve that point.

FIG. 2 schematically shows one row of a matrix (array) of photodetectors 10 (for example, photodiodes) arranged on the imaging surface 8 of the imaging device 7. Each photodetector 10 belonging to this row is connected to a common signal line 12 and is configured to output a detection signal generated by the light detection to the signal line 12.

Here, it is assumed that the direction of this row is the same as the direction in which the slit light is scanned by the optical force cutting method. That is, this is a case where the imaging device 7 is installed so that the row direction is parallel to the slit light scanning direction. In this case, the slit light image on the imaging surface 8 crosses each row of the matrix of the photodetectors 10 on the imaging surface 8. Here, the slit light scanning direction refers to the direction in which the slit light imaged on the imaging surface of the imaging device 7 moves in accordance with the slit light scanning by the deflection device 3.

[0030] The intensity pattern of the slit light incident at a certain moment with respect to the rows in which the groups of the photodetectors 10 are arranged is a pattern having one steep peak, for example, an incident intensity pattern 100 shown in the figure. It becomes. This is because one slit light crosses the line. Therefore, in a short time that the slit light can be considered to be stationary, the readout scan of the row, that is, the output of each photo detector 10 is applied to the one end force of the row in sequence to the other end of the signal line 12 one by one. As shown in the figure, an output signal 200 having one large peak flows through the signal line 12 (strictly speaking, the photodetectors 10 that output signals are sequentially turned off). The output signal on the signal line 12 does not become smooth as shown in the figure because it changes, and the general tendency is as shown in the figure). The illustrated example is an output signal 200 when the photodetector 10 is selected in order from left to right.

[0031] Then, if the time A from the start point power peak of one reading scan (that is, horizontal scanning (or row scanning)) or the time B from the peak to the end point of the reading scan is measured. The position of the peak in one line can be obtained.

[0032] The readout scanning for one row is performed by the horizontal scanning circuit 14. The horizontal scanning circuit 14 is composed of, for example, a parallel output shift register. In this case, one stage of the shift register is set to H (high) level (〃1〃) and the other stage is set to L (low) level (〃0〃), and this is the clock supplied from the timing control unit 16 According to the signal, one end force is shifted to the other end in order. Then, according to the output of each stage, the switch that intermittently connects between the output of each corresponding photodetector 10 and the signal line 12 is opened and closed. At the moment shown, only the stage corresponding to the fifth photodetector 10 from the left is at the high level, so the output of the fifth photodetector 10 (a low level signal close to 0) is output to the signal line 12. Is done.

[0033] By processing the signal on the signal line 12 by the position detection unit 18, the position of the peak of the incident light in the row at that time is obtained. The peak position can be obtained, for example, by counting the clock signal used as a reference for the shift operation of the horizontal scanning circuit 14 with a counter and obtaining the count value of the counter at the timing when the signal peak is detected. If the counter is cleared every horizontal scanning period, the counter value indicates the time A from the start of horizontal scanning to the peak detection time. The method for obtaining the peak position is not limited to this, and other examples will be described later.

With the configuration as described above, the peak position of incident light on a row is obtained. The recording processing unit 22 records this peak position in association with the count value of the counter 20 for obtaining the slit light projection angle. For example, the counter 20 may be incremented by one every horizontal scanning period of the horizontal scanning circuit 14, and the count value may be cleared every time scanning of the slit light is completed.

As described above, the force explaining the configuration of one row. If a plurality of rows are arranged, a pixel matrix (array) of the imaging surface 8 can be configured. In this case, one horizontal scanning circuit 14, timing control unit 16, counter 20 and recording processing unit 22 may be provided for the plurality of rows.

Further, in the above, the force that explained the processing in one horizontal scanning period If this processing is repeated during the scanning period of the slit light, the time at which the slit light is incident on each pixel position during the slit light scanning is calculated. You can ask for data. Thereby, basic data necessary for calculating the three-dimensional shape of the entire target object 4 can be collected. [0037] Note that, in the above, the description has been given of the case where the row direction of the pixel matrix is parallel to the slit scanning direction. In principle, if the image of the slit light crosses each row of the matrix, the above data collection processing can be executed. Therefore, when it is parallel, the force is the best mode, but it does not have to be parallel.

[0038] In the above, only one photodetector 10 is read out simultaneously in one row, but the outputs of a plurality of adjacent photodetectors 10 are read out simultaneously in order to improve sensitivity. I'll do it for you.

Next, an example of the circuit configuration of the three-dimensional measurement sensor of the present embodiment will be described with reference to FIG. Circuit elements corresponding to the components in the principle configuration of Fig. 2 are given the same symbols.

In the circuit configuration shown in FIG. 3, the pixel circuits 11 are arranged in a matrix, and a plurality of (four in the illustrated example) pixel circuits 11 belonging to the same row (horizontal line) are connected to a common signal line. Connected to 12. Although not shown, the pixel circuit 11 includes a photodetector 10 and a switch for switching between the output of the photodetector 10 and the signal line 12. This switch is turned on and off in response to a control signal supplied from the horizontal scanning circuit 14 via the address line 15. The switches of a plurality (four in the illustrated example) of pixel circuits 11 belonging to the same column are connected to a common address line 15 and turned on and off at the same time. The horizontal scanning circuit 14 supplies an ON signal (for example, an H level signal) only to the address line 15 of one column (or a predetermined number of consecutive columns) of the columns of the pixel circuit 11 group, and OFF to the other columns. Supply a signal (eg L level). Then, the horizontal scanning circuit 14 shifts the column to which the ON signal is supplied in one direction (for example, the right force is also to the left) by one column at regular intervals according to the clock signal supplied from the timing control unit 16. To go. By such a shift operation, during one horizontal running period, one row end force of the matrix and each column up to the other row end are selected in order. By such horizontal scanning, an output signal as illustrated in FIG. 2 indicating the position of the intensity peak of incident light in each row is output to the signal line 12 in each row.

Each signal line 12 is connected to the peak hold circuit 30, and the output of the peak hold circuit 30 is connected to the input of the timing detection circuit 32. These peak hold times A set of the path 30 and the timing detection circuit 32 corresponds to the position detection unit 18 in the principle configuration of FIG.

The peak hold circuit 30 holds and outputs the peak value of the signal input from the signal line 12. The output of each peak hold circuit 30 is input to the corresponding timing detection circuit 32. The timing detection circuit 32 generates a peak timing signal based on the output of the peak hold circuit 30, and generates and outputs a distance signal from the generated peak timing signal. The distance signal is the distance from the pixel in the row of pixels corresponding to the signal line 12 (ie, the pixel to which slit light is incident) to one end of the row (in this example, the right end). It is a signal which shows.

An example of the circuit configuration of the peak hold circuit 30 and the timing detection circuit 32 will be described with reference to FIG.

The peak hold circuit 30 in the illustrated example is a general peak hold circuit including two voltage follower circuits for impedance conversion of input and output and a capacitor C1 for holding a peak voltage. The signal line 12 is connected to the non-inverting input terminal of the operational amplifier OP 1 constituting the first voltage follower circuit. The output terminal of the operational amplifier OP1 is connected to the anode of the rectifying diode D1. The force sword side of the diode D1 is connected to the inverting input terminal of the operational amplifier OP1, thereby forming a first voltage follower circuit. The force sword side of the diode D1 is connected to the non-inverting input terminal of the op amp OP2 of the second voltage follower circuit. One end of a peak voltage holding capacitor C1 is connected to the non-inverting input terminal of the operational amplifier OP2 connected to the anode side of the diode D1, and the other end of the capacitor C1 is connected to the ground terminal. The output terminal of the operational amplifier OP2 is connected to the inverting input terminal of the operational amplifier OP2 itself, thereby forming a second voltage follower circuit. The output from the output terminal of the operational amplifier OP2 becomes the output of the peak hold circuit 30.

The timing detection circuit 32 includes a differentiation circuit 322 and a distance signal generation circuit 324.

The differentiation circuit 322 is a general differentiation circuit using an operational amplifier. One end of the capacitor C2 constituting the differentiation circuit 322 is connected to the output of the S peak hold circuit 30 and the other end is connected to the inverting input terminal of the operational amplifier OP3. Non-inverting input terminal of operational amplifier OP3 The child is connected to ground, and the output terminal is connected to the inverting input terminal of the operational amplifier OP3. A signal obtained by differentiating the output signal of the peak hold circuit 30 is output from the output terminal of the operational amplifier OP3.

[0047] The distance signal generation circuit 324 also includes a transistor TR1, a capacitor C3, a resistor Rl, and an operational amplifier OP4. The transistor TR1 in the example circuit is an NPN type, the base is connected to the output of the differentiating circuit 322, that is, the output terminal of the operational amplifier OP3, and the emitter is connected to the ground. The collector of transistor TR1 is connected to the non-inverting input terminal of operational amplifier OP4. One end of the resistor R1 is connected to the non-inverting input terminal of the operational amplifier OP4 connected to the collector of the transistor TR1, and the other end of the resistor R1 is connected to the horizontal scanning signal output terminal of the timing controller 16. The horizontal scanning signal is a signal indicating a horizontal scanning period, and is, for example, a signal that becomes H level during the horizontal scanning period. The non-inverting input terminal of the operational amplifier OP4 is connected to one end of a capacitor C3 that accumulates charges according to the horizontal scanning signal input via the resistor R1. The other end of the capacitor C3 is connected to ground. The output terminal of the operational amplifier OP4 is connected to the inverting input terminal of the operational amplifier OP4 itself to form a voltage follower circuit. The output signal from the output terminal of this operational amplifier OP4 is a distance signal indicating the peak position force of incident light in the pixel row and the distance to the end of the row.

Next, operations of the peak hold circuit 30 and the timing detection circuit 32 will be described with reference to the signal timing chart of FIG.

In FIG. 5, (a) is a timing chart of the horizontal scanning signal supplied from the timing control unit 16 to the distance signal generation circuit 324. This horizontal scanning signal becomes H level during the horizontal scanning period of the sensor, and is reset and becomes L level until the next horizontal scanning period starts.

(B) is a timing chart of an input signal to the peak hold circuit 30, that is, a signal on the signal line 12, and an output signal of the peak hold circuit 30. Since the photodetectors 10 that output optical detection signals by horizontal scanning are sequentially switched, the input signal from the signal line 12 becomes a signal having several peaks as shown by the solid line in the figure, and the peak of the incident light The maximum peak occurs at the timing when a signal is output from the photodetector 10 where the light is incident. Pea The output signal of the hold circuit 30 is a signal as shown by a broken line (it should be understood that the section where the broken line is not visible overlaps the solid line).

[0051] (c) is a timing chart of the output signal of the differentiating circuit 322. Since this signal is a signal obtained by differentiating the peak hold output signal indicated by the broken line in (b), the peak value of the input signal on the signal line 12 is held at the L level during the hold period. If the level of the input signal rises more than that, it becomes H level. The force level rises and falls several times before reaching the maximum peak of the input signal. After the maximum peak is exceeded, it is at the L level until the end of the horizontal running period.

[0052] (d) is a timing chart of the distance signal output from the distance signal generation circuit 324. The distance signal is a signal having a level corresponding to the charge accumulated in the capacitor C3 by the horizontal scanning signal. Here, when the transistor TR1 is turned on when the output signal of the differentiating circuit 322 (hereinafter referred to as differential output) becomes H level, the charge accumulated in the capacitor C3 is discharged to the ground through the transistor TR1. Therefore, every time the differential output becomes H level, the distance signal level is reset to the lowest value (0). However, after the input signal from the signal line 12 reaches the maximum peak, the differential output does not become H level until the end of the horizontal scanning period, so the charge is gradually accumulated in the capacitor C3, The level of the distance signal increases with time. Therefore, at the end of the horizontal scanning period, a distance signal of a level corresponding to the distance to the end position of the row (corresponding to time B in FIG. 2) is also output as the positional force of the photodetector 10 on which the slit light is incident. It will be.

Returning to FIG. 3 again, the distance signal output from the timing detection circuit 32 is converted into digital data by the AZD conversion 34. For example, if the AZD conversion has a bit number having a resolution equal to or higher than the number of pixels in one row, the incident position of the slit light on the row can be converted into data that can be discriminated with an accuracy higher than the pixel unit.

A latch (not shown) is provided at the subsequent stage of the AZD converter 34, and the digital data indicating the distance signal is latched by the latch circuit at the end timing of the horizontal scanning period. As a result, the latch at the end of each row holds digital data indicating the magnitude of the distance signal indicating the peak position of the incident light in the corresponding row at the end of the horizontal scanning period. In addition, at the start of the next horizontal scanning period, The latch value is cleared. In order to realize such an operation, the horizontal scanning signal is supplied from the timing control unit 16 to the latch circuit, for example, and the output of the AZD converter 34 is latched at the falling timing of the horizontal scanning signal, and at the rising timing. Clear the latched data.

The recording processing unit 22 shown in FIG. 2 sequentially reads out the data held by the latches of each row through the data output line 36 after the end of the horizontal scanning period and before the start of the next horizontal scanning period. . Then, for each latch of each row, the data obtained for the latch force (this indicates the peak position of the incident light in that row, that is, the highest luminance position), the number of that row, and the slit light projection at that time The combination with the count value of the counter 20 indicating the angle is stored in a predetermined storage device. If this operation is repeated for one scanning period of slit light, the data necessary to calculate the three-dimensional shape of the target object 4 is obtained. That is, the pixel position corresponding to the combination of the row number of each row obtained for each horizontal scanning period and the highest luminance position of each row (i.e., how many rows and columns) is specified, and the pixel corresponding to that pixel location The corresponding count value data may be adopted as the value of. As a result, it is possible to generate an image (referred to as a “projection angle image”) that indicates a count value (equivalent to the projection angle of the slit light) for each pixel on the imaging surface when that pixel reaches the maximum luminance. . In the subsequent processing device, the three-dimensional shape of the target object 4 can be calculated based on the projection angle image and the known arrangement position information of the light source and the imaging device.

Note that in the configuration of the position detection unit 18 including the peak hold circuit 30 and the timing detection circuit 32 described with reference to FIG. 4, the output signal of the peak hold circuit 30 can be simply input to the timing detection circuit 32. It can also be output as it is and used as a luminance signal. In this case, the luminance signal is converted into a digital value by an AZD converter (not shown) and then input to a latch circuit (not shown). That is, in this case, similarly to the AZD converter 34 and the latch circuit for the distance signal, the AZD converter and the latch circuit for the luminance signal are provided for each row. The latch circuit latches the output of the AZD change at the falling edge of the horizontal scanning signal and clears it at the next rising edge. The recording processing unit 22 stores the data (luminance value) held by the latch circuits in each row during a period in which the horizontal scanning signal is at the reset level (that is, from the end of the horizontal scanning period to the start of the next horizontal scanning period. (The period of time), and the corresponding row number and the count value of the counter 20 are stored in a predetermined storage device in association with each other. If this operation is repeated during the scanning period of the slit light, the maximum brightness value of each row at the count value timing is obtained for each count value of the counter 20.

[0057] Here, as described above, the recording processing unit 22 records the distance signal data (that is, the peak position of the incident light) of each row corresponding to each count value of the counter 20, so that each count value will be counted from now on. The position of the highest luminance in each row at the timing corresponding to is known. Therefore, it is possible to obtain a combination of the highest luminance value and the position on the row where the highest luminance value is obtained for each counter value timing (ie, for each projection angle of the slit light) and for each row. it can. Therefore, if the combination is collected over the scanning period of the slit light and mapped to the pixels in each row and each column of the pixel matrix, an image (referred to as a “luminance image”) corresponding to the maximum luminance of each pixel in that period is obtained. Obtainable. This image is close to an image obtained when the target object 4 is illuminated with a normal light source (not slit light).

In the above description, the data value of the distance signal and the data value of the luminance signal have been described as being individually recorded by the recording processing unit 22. However, these are associated with the count value of the counter 20 together. It is also preferable to record them.

Next, referring to FIG. 6, a configuration example of a measurement apparatus using the three-dimensional measurement sensor of the present embodiment will be described.

In the example of FIG. 6, the measuring device 100 includes a measuring circuit chip 110, an optical system 120, a scanning control device 130, a line light source 132, and a deflecting device 134. The optical system 120 is a system composed of optical components such as a lens, and forms an image of reflected light from the target object on the imaging surface of the measurement sensor 112. The measurement sensor 112 is a sensor including the circuit group illustrated in FIG. 3, that is, a matrix of pixel circuits 11, a horizontal scanning circuit 14, a peak hold circuit 30, a timing detection circuit 32, an A / D converter 34, and a latch circuit. is there. The processing circuit 114 is a circuit that provides the functions of the timing control unit 16, the counter 20, and the recording processing unit 22 in the configuration of FIG. 2 and the function for controlling the operation of the slit light. The processing circuit 114 outputs the above-described projection angle image (and a luminance image when a luminance image is generated) to the subsequent image capturing device 200.

[0061] These measurement sensor 112 and processing circuit 114 may be integrated on a one-chip. it can. Further, the measurement sensor 112 and the processing circuit 114 can be integrated into one chip, and the measurement circuit chip 100 can be configured.

The scanning control device 130 is a circuit that performs scanning control of slit light, and a linear light source that emits slit light in response to various control signals such as a horizontal scanning signal and a scanning clock signal supplied from the processing circuit 114. The on / off control of 132 and the mirror deflection angle in the deflecting device 134 are controlled.

The image capturing device 200 acquires a projection angle image and a luminance image from the processing circuit 114, and calculates the three-dimensional shape of the target object 4 from the projection angle image. The image capturing device 200 can display the calculated three-dimensional shape or luminance image on an attached display device or save it as a file.

[0064] The preferred embodiment of the present invention has been described above. As described above, the three-dimensional measurement sensor of the present embodiment can generate a projection angle image equivalent to the prior art disclosed in Patent Document 1 and Non-Patent Document 2 with a relatively small circuit scale. For example, in the circuit configuration of Non-Patent Document 2, a frame memory for 4 frames and a comparison circuit are required for each pixel, whereas in this embodiment, a peak hold circuit 30, a timing detection circuit 32, Since it is sufficient to provide the AZD conversion circuit 34 and the like, the circuit scale can be made smaller than that of Non-Patent Document 2.

Note that the circuit configurations shown in FIGS. 3, 4 and the like are merely examples, and various modifications are possible within the scope of the present invention.

For example, in the configuration shown in FIGS. 3, 4 and the like, the horizontal scanning signal is integrated to generate the distance signal, but this is merely an example. As an alternative configuration, there are configurations shown in FIGS. 7 and 8, respectively.

In the configuration of FIG. 7, waveform data equivalent to the time waveform of the distance signal generated by the distance signal generation circuit 324 of FIG. For the waveform data to be registered, the counter value for obtaining the peak position of incident light within the horizontal scanning period (that is, the elapsed time of the starting force in the horizontal scanning period) is plotted on the horizontal axis, and each counter value is Waves plotted with the value of the distance signal output at the end of the horizontal scanning period as the value on the vertical axis when the maximum peak of incident light is detected at the timing of the counter value Shape data.

In accordance with the horizontal scanning clock issued from the timing control unit 16, the value of the distance signal registered in the waveform memory 40 is sequentially output.

The analog signal is converted to an analog signal and input to the sample and hold circuit 44 in each row.

[0069] The sample hold circuit 44 is provided with a peak timing detection signal, that is, an output signal of the differentiating circuit 322 in the configuration of FIG. 4, as a control signal for controlling the sample / hold operation. Here, as shown in FIG. 5 (c), the output of the differentiation circuit 322 has a level corresponding to a small peak until the input signal from the signal line 12 reaches the maximum peak in the horizontal scanning period. Force that repeats up and down When the maximum peak is reached, the H level force After falling to the L level, the L level is maintained until the end of the horizontal scanning period. Therefore, if the sample-and-hold circuit 44 holds the signal input from the DZA conversion 42 at the falling edge of the rectangular pulse, the sample-and-hold circuit 44 starts from the end of the final horizontal scanning period. Thus, a distance signal having the same magnitude as that of the distance signal generation circuit 324 in FIG. 4 is output.

As described above, the configuration of FIG. 7 can be regarded as a configuration in which the distance signal generation circuit 324 in the configurations of FIGS. 3 and 4 is replaced with the waveform memory 40, the DZA converter 42, and the sample hold circuit 44. it can. The structure after the sample hold circuit 44 is omitted in illustration. This may be the same structure as that of the above embodiment described with reference to FIGS. It will be understood that processing similar to that in FIGS. 3 and 4 can be performed with such a configuration.

In FIG. 7, the level value of the distance signal read from the waveform memory 40 is converted into an analog signal and held by the sample hold circuit 44. The configuration of FIG. 8 shows the distance (that is, the peak position of incident light). The digital value is processed.

In the configuration of FIG. 8, instead of the distance signal generation circuit 324, the AZD converter 34, and the latch circuit (not shown) in the configurations of FIGS. 3 and 4, a counter 50, a lookup table (LUT) 52, and A digital memory 54 is provided.

In the configuration of FIG. 8, the counter 50 is reset at the start of the horizontal scanning period, and counts up the clock signal for horizontal scanning supplied from the timing control unit 16. The count value of this counter 50 is from the start of the horizontal scanning period to when the peak of incident light is detected. Indicates the time to a point, which is in principle proportional to the distance from the scanning start end of the row of the pixel matrix to the position of the peak of incident light in that row. Therefore, in principle, if the count value of the counter 50 is latched in the digital memory 54 at the fall timing of the peak timing detection signal, that is, the output of the differentiating circuit 322, the maximum input signal from the signal line 12 is obtained. Although the counter value held in the digital memory 54 may be updated several times above and below the differential output corresponding to the small peak up to the peak of, the end of the horizontal scanning period is not updated after the maximum peak. At that time, the counter value indicating the peak position of the incident light in the row in principle is held in the digital memory 54. Therefore, by reading the value held in the digital memory 54 of each row at the end of the horizontal scanning period, information on the highest luminance position of each row in that period can be obtained.

[0074] However, the elapsed time from the start point of the horizontal scanning period to the peak detection point of the incident light is not necessarily proportional to the distance from the scanning start side end of the row to the peak position of the incident light. . In addition, there may be individual differences for each measurement sensor in the relationship between the elapsed time and distance. Therefore, in the example of FIG. 8, a LUT 52 is provided to absorb such nonlinearity and individual differences in the relationship between elapsed time and distance. In the LUT 52, a value of “distance from the scanning start side edge of the row to the peak position of the incident light” corresponding to the count value of the counter 50 is registered. The LUT 52 outputs a distance value corresponding to the count value input from the counter 50. The value of this distance is latched in the digital memory 54 at the falling timing of the peak timing detection signal. By using the LUT 52 in this way, more accurate peak position information can be obtained.

[0075] It will be understood that processing similar to that in FIGS. 3 and 4 can be performed with such a configuration.

[0076] Note that the effect of calibrating individual differences and the like can be obtained by using the LUT 52 as described above. Even in the configuration of Fig. 7, the same calibration effect can be obtained by adjusting the waveform data registered in the waveform memory 40. Obtainable. In the sensor configuration in Figs. 3 and 4, the subsequent calculation unit that performs coordinate calculation based on trigonometry is provided with a lookup table that shows the relationship between the distance signal value and the actual peak position. ^ 34 digitized data If the coordinate calculation is performed after the correction by the calculation device, the same calibration is possible.

FIG. 9 shows still another example of the circuit configuration. In FIG. 9, the same components as those in FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.

In the circuit configuration of FIG. 9, the timing control horizontal scanning unit 14a is a combination of the horizontal scanning circuit 14 and the timing control unit 16 in the circuit configuration of FIG. Are displayed together.

[0079] The configuration of FIG. 9 includes the peak hold circuit 30, the timing detection circuit 32 in the configuration of FIG.

, An AZD converter 60, a data latch 62, a comparator 64, a data memory 66, and a counter circuit 68 are provided instead of the AZD converter 34 and the data latch. Of these, A / D converter 60

, Data latch 62, comparator 64 and data memory 66 for each signal line 12.

One by one.

[0080] The AZD converter 60 samples the signal flowing through the signal line 12 at every predetermined sample timing, converts it into a digital value, and outputs it. The output of the AZD translation 60 is input to the comparator 64 and the data latch 62. The comparator 64 compares the output of the A / D converter 60 with the digital value supplied from the data latch 62. As a result of comparison, if the former is greater than the latter, an H level signal is output. If the former is smaller than the latter, an L level signal is output. When an H level signal is input from the comparator 64, the data latch 62 latches the digital value supplied from the AZD conversion 60 at that time. In addition, when the comparator 64 force L level signal is input, the data latch 62 continues to hold the previously held digital value. The data latch 62 is reset after the end of one horizontal scanning period and before the start of the next horizontal scanning period. According to such an operation of the data latch 62, at a certain point in time in one horizontal scanning period, the data latch 62 is informed until the point of the signal flowing through the signal line 12 corresponding to the latch 62 (more strictly). Holds the maximum value (before the immediately preceding sample timing). Comparator 64 compares the maximum value held by data latch 62 with the value of the signal on signal line 12 at the current sample timing. Therefore, the comparator 64 outputs a signal that is almost the same as the differential output of FIG. 5C.

[0081] The counter circuit 68 counts the count values at regular intervals from the start to the end of one horizontal scanning period. Count up. The count interval of the count value is a period shorter than the time interval (that is, the time obtained by dividing the horizontal scanning period by the number of pixels in one line) that can identify individual pixel circuits 11 on one line. The count value of the counter circuit 68 is reset in accordance with timing control and a timing signal input from the horizontal scanning unit 14a between the end of one horizontal scanning period and the start of the next one horizontal scanning period. Therefore, the count value output from the counter circuit 68 is a value indicating the horizontal position on one line. The count value of the counter circuit 68 is supplied to the data memory 66 corresponding to each signal line 12. The data memory 66 latches the count value at the falling timing of the signal input from the comparator 64.

According to such a circuit configuration, at the end of one horizontal scanning period, each data memory 66 holds a count value corresponding to the incident position of the slit light on the corresponding signal line 12. That's true. The subsequent processing may be the same as that shown in FIG.

[0083] With this circuit configuration as well, the information of the three-dimensional shape can be obtained as in the configurations of FIGS.

FIG. 10 shows still another circuit configuration. In FIG. 10, elements corresponding to the elements shown in FIG.

In the circuit configuration of FIG. 9 above, the power of the signal line 12 digitalized by the AZD converter 60 and compared by the comparator 64 In the circuit configuration of FIG. 10, the comparison is analyzed. Run with. This will be described in detail below.

A signal flowing through the signal line 12 is input to the comparator 72 and the latch 70. The comparator 72 compares the signal on the signal line 12 with the signal supplied from the latch 70. If the former is larger than the latter as a result of the comparison, the comparator 72 compares the signal at the H level, and if the former is smaller than the latter, the comparator 72 compares the signal at the L level. The signal is output. When an H level signal is input from the comparator 72, the latch 70 holds the signal supplied from the signal line 12 at that time. Such a latch 70 can be configured as a sample and hold circuit, and the output of the comparator 72 serves as a trigger signal for instructing the sample operation for this circuit. When an L level signal is input from the comparator 72, the latch 70 continues to hold the signal level previously held and supplies the signal to the comparator 72. Also, latch 70 has 1 water It is reset after the end of the horizontal scanning period and before the start of the next horizontal scanning period.

Even with such a circuit configuration, a signal similar to the output of the comparator 64 in the circuit configuration of FIG. 9 can be supplied from the comparator 72 to the data memory 66. Therefore, as in FIG. 9, at the end of one horizontal scanning period, each data memory 66 holds a count value corresponding to the incident position of the slit light on the corresponding signal line 12. The subsequent processing may be the same as that shown in FIG. Therefore, with this circuit configuration as well, the information of the three-dimensional shape can be obtained in the same manner as the configurations of FIGS.

As described above, according to each embodiment of the present invention, equivalent data can be obtained with a simpler circuit configuration than the circuit configurations shown in Non-Patent Document 2 and Patent Document 1. A three-dimensional shape measurement sensor and a measurement apparatus using the same can be provided.

Claims

The scope of the claims

[1] A sensor for obtaining information for calculating a three-dimensional shape by a light section method, a photodetector array configured by arranging photodetectors in a matrix,

A row signal line provided for each row of the photodetector array and connected to the output of the photodetector belonging to the corresponding row;

The photodetectors of the photodetector array are selected in units of columns, the output signals of the selected photodetectors are output to the corresponding row signal lines, and the slit light to be scanned for the optical cutting method is selected. A row scanning circuit that sequentially switches a selected column from one end of the row to the other during a short row scanning period such that the image on the photodetector array can be considered substantially stationary;

A light receiving position detection circuit provided for each row signal line, and for each row scanning period, detects a peak in the row scanning period of a signal flowing through the corresponding row signal line, and based on the timing at which this peak is detected. A light receiving position detection circuit for obtaining information of the light receiving position in the row of the photodetector array in the row scanning period;

An output circuit that outputs information on the light receiving position obtained by each light receiving position detection circuit for each row scanning period; and

A three-dimensional measuring sensor.

[2] The counter further includes a counter that counts a clock signal that is an operation reference of the row scanning circuit, and each light receiving position detection circuit uses the count value of the counter at the timing when the peak is detected as information on the light receiving position. Output,

The three-dimensional measurement sensor according to claim 1.

[3] a counter that counts a clock signal that is an operation reference of the row scanning circuit;

A look-up table storing information on the light receiving position corresponding to the count value of the counter;

Further comprising

Each of the light receiving position detection circuits reads out and outputs the information of the light receiving position corresponding to the count value of the counter as well as the lookup table force.

The three-dimensional measurement sensor according to claim 1.

[4] The three-dimensional measurement according to claim 1, wherein each of the light reception position detection circuits generates a signal having a signal level corresponding to the light reception position as information on the reception position based on the timing at which the peak is detected. Sensor.

[5] Each of the light receiving position detection circuits

A peak hold circuit for holding the peak level of the signal on the corresponding row signal line;

A differentiating circuit for differentiating the output of the peak hold circuit;

An integrating circuit for integrating a predetermined reference signal;

A control element that discharges the charge accumulated in the integrating circuit when the output of the differentiating circuit is turned on, and accumulates charge in the integrating circuit while the output of the differentiating circuit is at the off level;

5. The three-dimensional measurement sensor according to claim 4, wherein the output circuit outputs an output of the integrating circuit when the row scanning period is completed as information on the light receiving position.

[6] The light receiving position detection circuit

A light receiving position signal generation circuit for generating a light receiving position signal having a signal level corresponding to the elapsed time from the start of the row scanning period;

A sample and hold circuit that is triggered by the output of the differentiating circuit and samples and holds the light receiving position signal at the time of the trigger;

5. The three-dimensional measurement sensor according to claim 4, wherein the output circuit outputs a signal held by the sample hold circuit when the row scanning period is completed as information on the light receiving position.

[7] An angle counter for obtaining the projection angle of the slit light on the measurement target;

The tertiary processing according to claim 1, further comprising: a storage processing unit that stores light reception position information obtained by each light reception position detection circuit in a row scanning period in association with a count value of the angle counter in the row scanning period. Former measuring sensor. An angle counter for obtaining the projection angle of the slit light with respect to the measurement target, information on the light receiving position obtained by each light receiving position detection circuit during the row scanning period, and an output of the peak hold circuit when the row running period is completed, A storage processing unit for storing in association with the count value of the angle counter in the row scanning period;

The three-dimensional measurement sensor according to claim 5 or 6, further comprising: